Particle Trajectory Representation Learning with Masked Point Modeling

TL;DR

This paper introduces PoLAr-MAE, a self-supervised learning method for LArTPC data that learns meaningful particle trajectories with high data efficiency, reducing the need for extensive labeled datasets.

Contribution

The paper presents a novel masked point modeling approach tailored for LArTPC images, achieving high performance with minimal labeled data and demonstrating emergent instance segmentation capabilities.

Findings

Achieves state-of-the-art segmentation with only 100 labeled events

Learns physically meaningful trajectory representations from unlabeled data

Releases a large dataset to facilitate further research

Abstract

Effective self-supervised learning (SSL) techniques have been key to unlocking large datasets for representation learning. While many promising methods have been developed using online corpora and captioned photographs, their application to scientific domains, where data encodes highly specialized knowledge, remains a challenge. Liquid Argon Time Projection Chambers (LArTPCs) provide high-resolution 3D imaging for fundamental physics, but analysis of their sparse, complex point cloud data often relies on supervised methods trained on large simulations, introducing potential biases. We introduce the Point-based Liquid Argon Masked Autoencoder (PoLAr-MAE), applying masked point modeling to unlabeled LArTPC images using domain-specific volumetric tokenization and energy prediction. We show this SSL approach learns physically meaningful trajectory representations directly from data. This yields remarkable data efficiency: fine-tuning on just 100 labeled events achieves track/shower semantic segmentation performance comparable to the state-of-the-art supervised baseline trained on $>$100,000 events. Furthermore, internal attention maps exhibit emergent instance segmentation of particle trajectories. While challenges remain, particularly for fine-grained features, we make concrete SSL's potential for building a foundation model for LArTPC image analysis capable of serving as a common base for all data reconstruction tasks. To facilitate further progress, we release PILArNet-M, a large dataset of 1M LArTPC events. Project site: https://youngsm.com/polarmae.